Determination of the inositol (1,4,5) trisphosphate requirement for histamine- and substance P-induced Ca2+ mobilisation in human U373 MG astrocytoma cells

Increasing the throughput of label-free cell assays to study the activation of G-protein-coupled receptors by using a serial agonist exposure protocol

Label-free, holistic assays, monitoring, for example, the impedance of cells on electrodes, are gaining increasing popularity in the evaluation of G-protein-coupled receptor (GPCR) ligands. It is the strength of these approaches to provide the integrated cellular response non-invasively, highly automated and with a device-dependent time resolution down to several milliseconds. With an increasing number of samples to be studied in parallel, the available time resolution is, however, reduced and the cost for the disposable sensor arrays may become limiting. Inspired by protocols from organ pharmacology, we investigated a simple serial agonist addition assay that circumvents these limitations in impedance-based cellular assays. Using a serial addition of increasing concentrations of a GPCR agonist while continuously monitoring the sample's impedance, we were able to establish a full concentration-response curve for the endogenous agonist histamine on a single layer of U-373 MG cells endogenously expressing the histamine 1 receptor (H1R). This approach is validated with respect to conventional, parallel agonist addition protocols and studies using H1R antagonists such as mepyramine. Applicability of the serial agonist addition assay was shown for other GPCRs known for their signaling via one of the canonical G-protein pathways, Gq, Gi/0 or Gs as well. The serial agonist addition protocol has the potential to further strengthen the output of label-free analysis of GPCR activation.

Activation of CD47 receptors causes proliferation of human astrocytoma but not normal astrocytes via an Akt-dependent pathway

CD47 is a membrane receptor that plays pivotal roles in many pathophysiological processes, including infection, inflammation, cell spreading, proliferation, and apoptosis. We show that activation of CD47 increases proliferation of human U87 and U373 astrocytoma cells but not normal astrocytes. CD47 function-blocking antibodies inhibit proliferation of untreated U87 and U373 cells but not normal astrocytes, suggesting that CD47 may be constitutively activated in astrocytoma. CD47 expression levels were similar in our three cell types. CD47 couples to G-proteins in astrocytes and astrocytoma and especially to the Gβγ dimer. Downstream signaling following CD47 activation involves Gβγ dimer-dependent activation of the PI3K/Akt pathway in astrocytoma cells but not in normal astrocytes. This pathway is known to be deregulated in astrocytoma, leading to cell proliferation and enhanced survival signals. Putative PLIC-1 interaction with CD47 in astrocytoma cells but not astrocytes may contribute to the proliferative effect observed upon activation of CD47. Our data indicate that CD47 receptors have a stimulatory role in cell proliferation and demonstrate for the first time that CD47 signals via the PI3K/Akt pathway in cancerous cells but not normal cells.

Histamine hyperpolarizes human glioblastoma cells by activating the intermediate-conductance Ca2+-activated K+ channel

The effects of histamine on the membrane potential and currents of human glioblastoma (GL-15) cells were investigated. In perforated whole cell configuration, short (3 s) applications of histamine (100 microM) hyperpolarized the membrane by activating a K(+)-selective current. The response involved the activation of the pyrilamine-sensitive H(1) receptor and Ca(2+) release from thapsigargin-sensitive intracellular stores. The histamine-activated current was insensitive to tetraethylammonium (3 mM), iberiotoxin (100 nM), and d-tubocurarine (100 microM) but was markedly inhibited by charybdotoxin (100 nM), clotrimazole (1 microM), and 1-[(2-chlorophenyl)diphenylmethyl]-1H-pyrazole (TRAM-34, 1 microM), a pharmacological profile congruent with the intermediate conductance Ca(2+)-activated K(+) (IK(Ca)) channel. Cell-attached recordings confirmed that histamine activated a K(+) channel with properties congruent with the IK(Ca) channel (voltage independence, 22 pS unitary conductance and slight inward rectification in symmetrical 140 mM K(+)). More prolonged histamine applications (2-3 min) often evoked a sustained IK(Ca) channel activity, which depended on a La(2+) (10 microM)-sensitive Ca(2+) influx. Intracellular Ca(2+) measurements revealed that the sustained IK(Ca) channel activity enhanced the histamine-induced Ca(2+) signal, most likely by a hyperpolarization-induced increase in the driving force for Ca(2+) influx. In virtually all cells examined we also observed the expression of the large conductance Ca(2+)-activated K(+) (BK(Ca)) channel, with a unitary conductance of ca. 230 pS in symmetrical 140 mM K(+), and a Ca(2+) dissociation constant [K(D(Ca))] of ca. 3 microM, at -40 mV. Notably in no instance was the BK(Ca) channel activated by histamine under physiological conditions. The most parsimonious explanation based on the different K(D(Ca)) for the two K(Ca) channels is provided.

Inositol 1,4,5-trisphosphate and ryanodine receptors mobilize calcium from a common functional pool in human U373 MG cells

This investigation concentrates on the change in Ca(2+) concentration ([Ca(2+)]) caused by ryanodine in U373 MG cells. This cell type from a human astrocytoma is a unique cellular model because it only expresses the type 3 ryanodine receptor (RyR3), which is generally the least abundant isoform. In the presence of physiological [Ca(2+)] in the extracellular medium, U373 MG cells are caffeine-insensitive, even after forskolin treatment, and ryanodine-sensitive only when an unusually high concentration (30 microM) is applied. Xestospongin C behaves like thapsigargin and therefore cannot be used as a selective antagonist of inositol 1,4,5-trisphosphate receptors (InsP(3)Rs). After ryanodine challenge, addition of an analog of Substance P (SP), which should deplete InsP(3)-sensitive stores, has no effect on [Ca(2+)](i). After thapsigargin treatment, which unmasks the calcium leak from intracellular stores, neither ryanodine nor SP change [Ca(2+)](i), suggesting that thapsigargin completely depletes the ryanodine-sensitive and the InsP(3)-sensitive stores of U373 MG cells. Finally, in experiments monitoring the [Ca(2+)] in intracellular stores, InsP(3) stimulation of permeabilized cells causes a decrease in [Ca(2+)] that is not affected by subsequent ryanodine treatment. Our results support the conclusion that U373 MG cells express both InsP(3)Rs and RyRs that can individually or in combination mobilize only one functional Ca(2+) pool.

Histamine potentiates IP(3)-mediated Ca(2+) release via thapsigargin-sensitive Ca(2+) pumps

We have studied the histamine-induced potentiation of inositol 1,4,5-trisphosphate (IP(3))-mediated Ca(2+) release in HeLa cells. Intracellular IP(3) levels were increased by IP(3) dialysis with the whole-cell configuration of the patch-clamp technique (cell dialysis of IP(3)). Low concentrations of extracellular histamine (1 microM) accelerated the rate of IP(3)-mediated Ca(2+) release, an effect that required the coincidence of both histamine signalling and the increase in IP(3) levels. Our data suggest that the potentiation effect of histamine cannot be explained simply by agonist-induced increase in IP(3) levels. Disordering microfilaments with cytochalasin D and microtubules with colchicine caused a decrease in the histamine-induced Ca(2+) response. Furthermore, both cytochalasin D and colchicine diminished the rate of IP(3)-mediated Ca(2+) release, while only the former reduced slightly the histamine-induced potentiation effect. Remarkably, rapid inhibition of SERCA pumps with thapsigargin to avoid the depletion of internal Ca(2+) stores diminished the histamine-induced potentiation of IP(3)-mediated Ca(2+) release, without affecting the rate of IP(3)-mediated Ca(2+) release. These data indicate that histamine-induced potentiation of Ca(2+) release in HeLa cells requires active SERCA pumps and suggest that SERCA pumps are an important factor in determining the efficiency of agonist-induced Ca(2+) release.

Thiomersal enhances the binding of histamine to the H1 receptor, but not histamine-stimulated inositol phosphate formation

Thiomersal (thimerosal) was a weak inhibitor of the binding of [(3)H]mepyramine to histamine H(1) receptors in guinea-pig cerebellar membranes (11 +/- 1% inhibition at 10 microM, 32 +/- 3% inhibition at 300 microM). However, in the concentration range 3-30 microM, thiomersal enhanced the binding of histamine to the H(1) receptor, as reflected by the displacement of curves of histamine inhibition of [(3)H]mepyramine binding to lower concentrations, without any change in the Hill coefficient. The ratio of the IC50 values (the concentration giving 50% inhibition) in the absence and presence of thiomersal increased from 1.8 with 3 microM to 3.6 with 30 microM thiomersal. There was no consistent effect of thiomersal at concentrations of 30 microM and below on curves of mepyramine inhibition of [(3)H]mepyramine binding. In the presence of 10 microM thiomersal histamine-induced accumulation of inositol phosphates in U373 MG astrocytoma cells was partially inhibited (37 +/- 8% inhibition of the maximum response), without any significant change in the EC50 (the concentration giving the half maximal response) for histamine. Thus although histamine binding was potentiated by thiomersal, there was no potentiation of an H(1) receptor-mediated functional response.

Luminal Ca(2+) and the activity of sarcoplasmic reticulum Ca(2+) pumps modulate histamine-induced all-or-none Ca(2+) release in smooth muscle cells

We have studied histamine (HA)-evoked intracellular Ca(2+) release in single, freshly isolated myocytes from the guinea pig urinary bladder. Short applications of histamine (5 s) produced a thapsigargin (TG)-sensitive transient increase in intracellular calcium concentration ([Ca(2+)](i)). It was established that histamine and caffeine (Caff) released Ca(2+) from the same intracellular stores in these cells. Reducing the Ca(2+) content of internal stores by incubating cells with U-73343 or cyclopiazonic acid (CPA) inhibited the histamine-evoked Ca(2+) release in 69% and 60% of cells, respectively. Under these conditions, all cells released Ca(2+) in response to either caffeine or acetylcholine (ACh). However, decreasing internal Ca(2+) stores by removing external Ca(2+) inhibited histamine-induced Ca(2+) mobilization in only 22% of cells. A similar small fraction of cells was inhibited when sarcoplasmic reticulum (SR) Ca(2+) pumps were quickly blocked to avoid a significant reduction of luminal Ca(2+). In conclusion, lowering the luminal Ca(2+) content in combination with an impairment of the SR Ca(2+) pump activity significantly diminishes the ability of histamine to evoke an all-or-none intracellular Ca(2+) release.

The initial inositol 1,4,5-trisphosphate response induced by histamine is strongly amplified by Ca(2+) release from internal stores in smooth muscle

We have studied the Ca(2+)-dependence and wortmannin-sensitivity of the initial inositol 1,4,5-trisphosphate (Ins(1,4,5)P(3)) response induced by activation of either histamine or muscarinic receptors in smooth muscle from guinea pig urinary bladder. Activation of H(1) receptors with histamine (100 microM) produced a significant elevation in Ins(1,4,5)P(3) levels with only 5s stimulation and in the presence of external Ca(2+). However, this response was abolished fully by either the prolonged absence of external Ca(2+) or the depletion of internal Ca(2+) stores with thapsigargin (100nM) or ryanodine (10 microM). In contrast, the same conditions only slightly reduced the initial Ins(1,4,5)P(3) response induced by carbachol. The prolonged incubation of smooth muscle in 10 microM wortmannin to inhibit type III PI 4-kinase abolished both the early histamine-evoked Ins(1,4,5)P(3) and Ca(2+) responses. Conversely, wortmannin did not alter Ca(2+) release induced by carbachol, despite a partial reduction of its Ins(1,4,5)P(3) response. Collectively, these data indicate that the detectable histamine-induced increase in Ins(1,4,5)P(3) is more the consequence of Ca(2+) release from internal stores than a direct activation of phospholipase C by H(1) receptors. In addition, the effect of wortmannin implies the existence of a Ca(2+)-dependent amplification loop for the histamine-induced Ins(1,4,5)P(3) response in smooth muscle.

Histamine H1 receptor activation stimulates mitogenesis in human astrocytoma U373 MG cells

In human astrocytoma U373 MG cells that express histamine H1 receptors (180 +/- 6 fmol/mg protein) but not H2 or H3 receptors, histamine stimulated mitogenesis as assessed by [3H]-thymidine incorporation (173 +/- 2% of basal; EC50, 2.5 +/- 0.4 microM). The effect of 100 microM histamine was fully blocked by the selective H1 antagonist mepyramine (1 microM) and was markedly reduced (93 +/- 4% inhibition) by the phospholipase C inhibitor U73122 (10 microM). The activator of protein kinase C (PKC) phorbol 12-tetradecanoyl-13-acetate (TPA, 100nM) stimulated [3H]-thymidine incorporation (270 +/- 8% of basal), and this response was not additive with that to 100 microM histamine. The incorporation of [3H]-thymidine induced by 100 microM histamine was partially reduced by the PKC inhibitor Ro 31-8220 (57 +/- 7% inhibition at 300 nM) and by the compound PD 098,059 (30 microM, 62 +/- 14% inhibition), an inhibitor of the mitogen-activated kinase (MAPK) kinases MEK1/MEK2. These results show that histamine H1 receptor activation stimulates the proliferation of human astrocytoma U373 MG cells. The action of histamine appears to be partially mediated by PKC stimulation and MAPK activation.

Inositol 1,4,5-trisphosphate-independent calcium signalling by platelet-derived growth factor in the human SH-SY5Y neuroblastoma cell

In adherent SH-SY5Y human neuroblastoma cells, activation of G-protein-coupled muscarinic M3 receptors evoked a biphasic elevation of both intracellular [Ca(2+)] ([Ca(2+)]i) and inositol-1,4,5-trisphosphate (D-Ins(1,4,5)P3) mass. In both cases, temporal profiles consisted of rapid transient elevations followed by a decline to a lower, yet sustained level. In contrast, platelet-derived growth factor (PDGF), a receptor tyrosine kinase agonist acting via PDGF receptor b chains in these cells, elicited a slow and transient elevation of [Ca(2+)]i that returned to basal levels within 5 to 10 min with no evidence of inositol phosphate generation. Full responses for either receptor type required intracellular and extracellular Ca(2+) and mobilization of a shared thapsigargin-sensitive intracellular Ca(2+) store. Strategies that affected the ability of D-Ins(1,4,5)P3 to interact with the Ins(1,4,5)P3-receptor demonstrated an Ins(1,4,5)P3-dependency of the muscarinic receptor-mediated elevation of [Ca(2+)]i but showed that PDGF-mediated elevations of [Ca(2+)]i are Ins(1,4,5)P3-independent in these cells.

Enhanced inducible mGlu1alpha receptor expression in Chinese hamster ovary cells

Inducible expression of the group-I metabotropic glutamate receptor (mGlu1alpha) in Chinese hamster ovary cells allows for the study of receptor density dependent effects. However, expression levels attainable with this system are lower than those reported for various brain regions and achieved by conventional (constitutive) transfection. Thus, direct comparison of mGlu1alpha receptor-mediated responses in this inducible expression system with those for receptors expressed heterologously or in vivo is compounded. We show here that inducible expression can be selectively augmented by butyrate pretreatment to levels approaching those reported for cerebral tissue. Enhanced mGlu1alpha receptor protein levels, agonist-induced inositol phosphate accumulation, as well as single-cell inositol 1,4,5-trisphosphate production and intracellular Ca(2+) mobilization occurred following co-induction with butyrate. In contrast, endogenous purinoceptor function was unaffected. Importantly, the ability to titrate receptor expression by varying isopropyl beta-thiogalactoside concentration was retained. Sodium butyrate thus offers a simple and convenient method to enhance inducible gene expression to levels found in vivo.

Reassessment of the Ca2+ sensing property of a type I metabotropic glutamate receptor by simultaneous measurement of inositol 1,4,5-trisphosphate and Ca2+ in single cells

Transient transfection of Chinese hamster ovary or baby hamster kidney cells expressing the Group I metabotropic glutamate receptor mGlu1alpha with green fluorescent protein-tagged pleckstrin homology domain of phospholipase Cdelta1 allows real-time detection of inositol 1,4,5-trisphosphate. Loading with Fura-2 enables simultaneous measurement of intracellular Ca(2+) within the same cell. Using this technique we have studied the extracellular calcium sensing property of the mGlu1alpha receptor. Quisqualate, in extracellular medium containing 1.3 mm Ca(2+), increased inositol 1,4,5-trisphosphate in all cells. This followed a typical peak and plateau pattern and was paralleled by concurrent increases in intracellular Ca(2+) concentration. Under nominally Ca(2+)-free conditions similar initial peaks in inositol 1,4,5-trisphosphate and Ca(2+) concentration occurred with little change in either agonist potency or efficacy. However, sustained inositol 1,4,5-trisphosphate production was substantially reduced and the plateau in Ca(2+) concentration absent. Depletion of intracellular Ca(2+) stores using thapsigargin abolished quisqualate-induced increases in intracellular Ca(2+) and markedly reduced inositol 1,4,5-trisphosphate production. These data suggest that the mGlu1alpha receptor is not a calcium-sensing receptor because the initial response to agonist is not sensitive to extracellular Ca(2+) concentration. However, prolonged activation of phospholipase C requires extracellular Ca(2+), while the initial burst of activity is highly dependent on Ca(2+) mobilization from intracellular stores.

Single-cell imaging of graded Ins(1,4,5)P3 production following G-protein-coupled-receptor activation

The pleckstrin homology domain of phospholipase Cdelta1 (PH(PLCdelta)) binds Ins(1,4,5)P(3) and PtdIns(4,5)P(2) specifically, and can be used to detect changes in Ins(1,4,5)P(3) in single cells. A fusion construct of PH(PLCdelta) and enhanced green fluorescent protein (EGFP-PH(PLCdelta)) associates with the plasma membrane due to its association with PtdIns(4,5)P(2). However, PH(PLCdelta) has greater affinity for Ins(1,4,5)P(3) than PtdIns(4,5)P(2), and translocates to the cytosol as Ins(1,4,5)P(3) levels rise. Prolonged activation of group I metabotropic glutamate receptor 1alpha expressed in Chinese-hamster ovary cells or endogenous M(3) muscarinic receptors in SH-SY5Y neuroblastoma cells gave an initial transient peak in translocation, followed by a sustained plateau phase. This closely followed changes in cell population Ins(1,4,5)P(3) mass, but not PtdIns(4,5)P(2) levels, which decreased monophasically, as determined by radioreceptor assay. Translocation thus provides a real-time method to follow increases in Ins(1,4,5)P(3). Graded changes in Ins(1,4,5)P(3) in Chinese-hamster ovary-lac-mGlu1alpha cells could be detected with increasing glutamate concentrations, and dual loading with fura 2 and EGFP-PH(PLCdelta) showed that changes in intracellular Ca(2+) concentration closely paralleled Ins(1,4,5)P(3) production. Moreover, Ins(1,4,5)P(3) accumulation and intracellular Ca(2+) mobilization within single cells is graded in nature and dependent on both agonist concentration and receptor density.

Intracellular sphingosine 1-phosphate production: a novel pathway for Ca2+ release

Sphingolipids such as sphingosine 1-phosphate (SPP) and sphingosylphosphorylcholine have long been recognized to possess Ca2+ mobilizing activity, yet to date little is known about their mechanism of action, or indeed their significance as Ca2+ mobilizing intracellular messengers. The recent discovery of extracellular receptors for the sphingolipids has further complicated the interpretation of many experiments in this field. This paper reviews the current literature in which molecular and pharmacological approaches have begun to uncover the signalling components associated with intracellular SPP production and Ca2+ mobilization. The functional significance of this novel Ca2+ release pathway is also discussed.

Lysophosphatidic acid-induced Ca2+ mobilization requires intracellular sphingosine 1-phosphate production. Potential involvement of endogenous EDG-4 receptors

Lysophosphatidic acid (LPA)-mediated Ca(2+) mobilization in human SH-SY5Y neuroblastoma cells does not involve either inositol 1,4, 5-trisphosphate (Ins(1,4,5)P(3))- or ryanodine-receptor pathways, but is sensitive to inhibitors of sphingosine kinase. This present study identifies Edg-4 as the receptor subtype involved and investigates the presence of a Ca(2+) signaling cascade based upon the lipid second messenger molecule, sphingosine 1-phosphate. Both LPA and direct G-protein activation increase [(3)H]sphingosine 1-phosphate levels in SH-SY5Y cells. Measurements of (45)Ca(2+) release in premeabilized SH-SY5Y cells indicates that sphingosine 1-phosphate, sphingosine, and sphingosylphosphorylcholine, but not N-acetylsphingosine are capable of mobilizing intracellular Ca(2+). Furthermore, the effect of sphingosine was attenuated by the sphingosine kinase inhibitor dimethylsphingosine, or removal of ATP. Confocal microscopy demonstrated that LPA stimulated intracellular Ca(2+) "puffs," which resulted from an interaction between the sphingolipid Ca(2+) release pathway and Ins(1,4,5)P(3) receptors. Down-regulation of Ins(1,4,5)P(3) receptors uncovered a Ca(2+) response to LPA, which was manifest as a progressive increase in global cellular Ca(2+) with no discernible foci. We suggest that activation of an LPA-sensitive Edg-4 receptor solely utilizes the production of intracellular sphingosine 1-phosphate to stimulate Ca(2+) mobilization in SH-SY5Y cells. Unlike traditional Ca(2+) release processes, this novel pathway does not require the progressive recruitment of elementary Ca(2+) events.

Characteristics of the Ca(2+)-dependent inhibition of cyclic AMP accumulation by histamine and thapsigargin in human U373 MG astrocytoma cells

1. Histamine, acting on H(1)-receptors, caused a Ca(2+)-dependent inhibition of forskolin- and isoprenaline-induced cyclic AMP accumulation in monolayers of human U373 MG cells (IC(50) 1.3+/-0.3 microM, maximum inhibition 66+/-3%). The inhibition was not reversed by the protein kinase inhibitor K-252A. 2. Thapsigargin also inhibited cyclic AMP accumulation (IC(50) 6.0+/-0.3 nM, maximum inhibition 72+/-1%). In the absence of extracellular Ca(2+) 5 microM thapsigargin caused only a 12+/-2% inhibition of cyclic AMP accumulation. 3. The inhibitory effect of 100 nM thapsigargin on forskolin-stimulated cyclic AMP accumulation was blocked by La(3+) (best-fit maximum inhibition 81+/-4%, IC(50) 125+/-8 nM). In contrast, the inhibitory action of 10 microM histamine was much less sensitive to reversal by 1 microM La(3+) (33+/-5% reversal, compared with 78+/-6% reversal of the inhibition by thapsigargin measured concurrently). However, in the presence of both thapsigargin and histamine the inhibition of cyclic AMP accumulation was reversed by 1 microM La(3+) to the same extent as the inhibition by thapsigargin alone. 4.++Thapsigargin (5 microM)+1 microM La(3+) caused only a 20+/-1% inhibition of histamine-stimulated phosphoinositide hydrolysis. 5. There was no indication from measurement of intracellular Ca(2+) of any persistent La(3+)-insensitive Ca(2+) entry component activated by histamine. 6. The results provide evidence that Ca(2+) entry is required for the inhibition by histamine and thapsigargin of drug-induced cyclic AMP accumulation in U373 MG astrocytoma cells. The differential sensitivity of the inhibitory action of the two agents to block by La(3+) suggests that more than one pathway of Ca(2+) entry is involved.

Extracellular calcium-sensing receptor induces cellular proliferation and activation of a nonselective cation channel in U373 human astrocytoma cells

A receptor for extracellular calcium ions (Ca2+o), cloned from parathyroid gland, serves a critical function in Ca2+o homeostasis by regulating PTH release via "sensing" of its physiological agonist, Ca2+o. Its cloning from rat striatum revealed that the extracellular calcium-sensing receptor (CaR) could be involved in sensing ambient Ca2+o within the brain, where Ca2+ plays key roles in virtually all aspects of central nervous system (CNS) function. The CaR is expressed in neurons, oligodendrocytes, microglia and the human astrocytoma cell line, U87 where its functions include control of cellular proliferation and modulation of ion channels, such as outward K+ channels and nonselective cation channels (NCC). In this report, we have shown that the CaR is expressed in U373 cells as assessed by RT-PCR using CaR-specific primers followed by sequencing of the amplified products, by Northern blot analysis using a CaR-specific probe as well as by Western analysis utilizing a specific polyclonal anti-CaR antiserum. Furthermore, agents known to activate the cloned CaR induce increases in cellular proliferation and the open probability of an NCC. Thus our study strongly suggests that elevated levels of Ca2+o, acting via the CaR, activate an NCC that could contribute to the associated CaR-induced stimulation of proliferation.

Glutamate-triggered calcium signalling in mouse bergmann glial cells in situ: role of inositol-1,4,5-trisphosphate-mediated intracellular calcium release

The mechanisms of glutamate-induced changes in intracellular free calcium concentration in Bergmann glial cells in mouse cerebellar slices were investigated by Fura-2-based microfluorimetry. Extracellular applications of glutamate, quisqualate and kainate triggered an increase in cytoplasmic calcium concentration, whereas N-methyl-D-aspartate and alpha-amino-3-hydroxy-5-methylisoxazole-4-propionate were ineffective. The calcium elevation triggered by kainate was completely blocked by removal of calcium ions from the external solutions or by slice incubation with 6-cyano-7-nitroquinoxaline-2,3-dione. Conversely, both glutamate- and quisqualate-induced intracellular calcium transients were only slightly attenuated by slice incubation with either 6-cyano-7-nitroquinoxaline-2,3-dione or calcium-free solution, suggesting the intracellular origin for calcium ions. The glutamate-triggered cytosolic calcium increases were inhibited by slice incubation with thapsigargin, the inhibitor of intracellular calcium pumps, or by intracellular perfusion of Bergmann glial cells with heparin, the antagonist of inositol-1,4,5-trisphosphate-gated calcium release channels. Therefore the calcium release from inositol-1,4,5-trisphosphate-sensitive intracellular stores plays the major role in glutamate-induced calcium signalling. We concluded that Bergmann glial cells express calcium permeable ionotropic glutamate receptors, which might be important for generation of fast calcium signals. However, slow glutamate-evoked calcium signals are mostly determined by inositol-1,4,5-trisphosphate-dependent intracellular signalling chain.